Abstract
We evaluated the performance characteristics of chromID CARBA and HardyCHROM Carbapenemase for the detection of carbapenemase-producing Enterobacteriaceae (CPE). A CPE prevalence study was conducted using chromID CARBA; this demonstrated that in low-prevalence settings, CPE screening agars may lack specificity, and confirmation of putative isolates is necessary.
TEXT
Over the last decade, carbapenem-resistant Enterobacteriaceae (CRE) have disseminated worldwide (1–4). Carbapenem resistance can be mediated by multiple mechanisms, including enzymes that can inactivate carbapenems (carbapenemases) or the presence of efflux or porin mutations. Perhaps the most alarming are the carbapenemase-producing isolates, which are frequently housed on mobile genetic elements, are resistant to most (if not all) β-lactams, and are often resistant to multiple other classes of drugs as well.
Due to the lack of therapeutic options for these organisms, tremendous efforts have been focused on preventing the spread of carbapenemase-producing Enterobacteriaceae (CPE). In order to identify and isolate asymptomatic carriers, active surveillance for fecal carriage in hospitalized patients is becoming increasingly common.
The objective of this study was to compare the limit of detection (LOD) of two commercially available chromogenic agars, chromID CARBA (bioMérieux, Durham, NC) and HardyCHROM Carbapenemase (recently renamed HardyCHROM CRE; Hardy Diagnostics, Santa Maria, CA), for the recovery of CPE from clinical specimens. In addition, a prevalence study was conducted to determine the rate of fecal and respiratory carriage of CPE at Barnes-Jewish Hospital (St. Louis, MO), a tertiary care academic medical center.
(This work was presented in part at the 114th General Meeting of the American Society for Microbiology, Boston, MA, 17 to 20 May 2014.)
For the LOD analysis, CPE-negative remnant stool specimens from Clostridium difficile testing were pooled, divided into 5-ml aliquots, and frozen at −20°C until use. Bacteria were inoculated into stool specimens, resulting in a series of specimens with a final concentration of 101 to 104 CFU/ml. To mimic a rectal swab specimen, an ESwab (Copan Diagnostics, Murrieta, CA) was placed into the stool and subsequently into ESwab transport media. After 1 h at room temperature, the swab was vortexed, and 1 drop of the specimen was inoculated onto chromID CARBA and HardyCHROM Carbapenemase agars. After 24 h, the plates were evaluated, and LOD was determined for each isolate (n = 12) (Table 1). A total of three replicates were performed.
TABLE 1.
Limits of detection for chromID CARBA and HardyCHROM carbapenemase media using spiked stool specimensa
| Strain | Limit of detection (CFU/ml) with direct inoculation |
Limit of detection (CFU/ml) with CDC method |
||
|---|---|---|---|---|
| ChromID CARBA | HardyCHROM Carbapenemase | ChromID CARBA | HardyCHROM | |
| KPC isolates | ||||
| Klebsiella pneumoniae BAA-1705 | 103 | 104 | Not recovered | Not recovered |
| Klebsiella pneumoniae | 102 | 103 | Not recovered | Not recovered |
| Escherichia coli 1 | 103 | Not recovered | Not recovered | Not recovered |
| Escherichia coli 2 | 103 | 103 | 104 | 104 |
| Enterobacter cloacae complex 1 | 103 | Not recovered | Not recovered | Not recovered |
| Enterobacter cloacae complex 2 | 102 | Not recovered | Not recovered | Not recovered |
| Citrobacter freundii complex 1 | 103 | Not recovered | Not recovered | Not recovered |
| Citrobacter freundii complex 2 | 102 | Not recovered | Not recovered | Not recovered |
| NDM-1 isolates | ||||
| Klebsiella pneumoniae BAA-2146 | 102 | 102 | 102 | 102 |
| Escherichia coli | 103 | 103 | Not recovered | Not recovered |
| Enterobacter cloacae complex | 102 | 103 | Not recovered | Not recovered |
| Providencia rettgeri | Not recovered | Not recovered | Not recovered | Not recovered |
Tested serial dilutions, 101 to 104 CFU/ml.
For comparison, 1 drop of the ESwab specimen was also added to 5 ml of tryptic soy broth (TSB) containing a 10-μg meropenem disk (BD Diagnostics, Franklin Lakes, NJ; CDC method) (5). The next day, 1 drop of the TSB was inoculated onto chromID CARBA and HardyCHROM Carbapenemase agars. After 24 h, the plates were examined, and the LOD was determined for each isolate. Two replicates were performed. A third replicate was performed using a larger volume, 100 μl of TSB, which was inoculated onto both chromogenic agars.
The Barnes-Jewish Hospital prevalence study was carried out over 12 weeks, from October 2013 to January 2014. This study was approved by the Washington University School of Medicine Human Research Protection Office. Remnant stool and rectal swab specimens submitted for C. difficile testing, C. difficile testing plus enteric culture, or vancomycin-resistant enterococci (VRE) screening, as well as tracheal aspirate specimens, were deidentified by an honest broker. A 10-μl loop of specimen was inoculated onto chromID CARBA agar, with the exception of VRE swabs, which were inoculated directly onto the agar. Multiple lot numbers of chromID CARBA were used. The plates were examined for growth after 24 and 48 h of incubation. All growth was subcultured to blood agar plates (BD Diagnostics) and incubated overnight. The next day, isolates were Gram stained and subsequently analyzed by matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) on the Vitek MS (IVD v.2.0; bioMérieux) (6–9). If identification was not achieved, the isolates were analyzed by the Bruker Biotyper (library version 3.4; Bruker Daltonics, Billerica, MA) (10, 11).
A subset of 60 specimens from the prevalence study, submitted for C. difficile testing, were inoculated onto both chromID CARBA and HardyCHROM Carbapenemase media, as described above, for direct comparison of the two media.
Antimicrobial susceptibility testing was performed on all Gram-negative organisms recovered on chromID CARBA agar using Kirby-Bauer disk diffusion. Categorical interpretation of zone sizes for all antimicrobials was based on the 2013 Clinical and Laboratory Standards Institute (CLSI) disk diffusion criteria (12). Although carbapenem disk diffusion breakpoints do not exist for nonfermenting Gram-negative bacilli, disk diffusion was used as an initial screen for carbapenem resistance. All Enterobacteriaceae and Pseudomonas putida isolates and carbapenem-nonsusceptible Pseudomonas aeruginosa isolates were tested for the blaKPC (13), blaNDM-1 (14), blaIMP, and blaVIM (15) genes by real-time PCR on the Cepheid SmartCycler II (Cepheid, Sunnyvale, CA). In addition, all carbapenem-nonsusceptible Enterobacteriaceae were tested for the presence of the blaCTX-M gene using conventional PCR detected by agarose gel electrophoresis (15).
Six (50%) of the 12 isolates did not grow at the highest concentration tested (104 CFU/ml) on HardyCHROM carbapenemase agar, including Klebsiella pneumoniae carbapenemase (KPC)-producing strains of Escherichia coli and Citrobacter freundii complex, whereas a single isolate did not grow on the chromID CARBA agar (a Providencia rettgeri NDM-1–producing isolate). For the isolates that did grow, the limit of detection was determined to be 102 to 104 CFU/ml (Table 1).
Based on our initial findings demonstrating a superior LOD, chromID CARBA was selected for the prospective prevalence study. A total of 611 specimens, including 561 fecal samples and 50 tracheal aspirates, were inoculated directly onto chromID CARBA agar. Of the 561 fecal specimens, 450 were submitted for C. difficile toxin testing and 111 were submitted for VRE screening. Forty-nine of the isolates submitted for C. difficile testing were also submitted for enteric culture. The specimens represented multiple patient age groups and hospital units, including intensive care and oncology.
Of the 611 specimens inoculated onto chromID CARBA, 139 isolates were recovered from 131 specimens. The breakdown of organisms recovered is shown in Table 2. Of the 9 Enterobacteriaceae isolates, 6 were recovered after 24 h of incubation, and 3 isolates were recovered after 48 h of incubation. After 24 h, the overall specificity was 70%, and the specificity was 95.3% based upon interpretation of the chromogenic medium.
TABLE 2.
Isolates recovered on chromID CARBA agar after 24 and 48 h of incubation
| Isolate | No. of isolates after: |
Total no. of isolates | No. of PCR-positive isolatesa | |
|---|---|---|---|---|
| 24 h | 48 h | |||
| Aeromonas sp. | 1 | 0 | 1 | NA |
| Burkholderia multivorans | 0 | 1 | 1 | NA |
| Candida sp. | 1 | 35 | 36 | NA |
| Citrobacter freundii complex | 1 | 1 | 2 | 0/2 |
| Enterobacter sp. | 2 | 1 | 3 | 0/3 |
| Escherichia coli | 1 | 0 | 1 | 0/1 |
| Enterococcus sp. | 1 | 22 | 23 | NA |
| Klebsiella oxytoca | 1 | 1 | 2 | 1/2 |
| Pseudomonas aeruginosa | 27 | 23 | 50 | 0/10 |
| Pseudomonas sp. | 3 | 6 | 9 | 0/3 |
| Stenotrophomonas maltophilia | 4 | 6 | 10 | NA |
| Serratia marcescens | 1 | 0 | 1 | 0/1 |
| Total | 43 | 96 | 139 | 1/22 |
Isolates were tested for the presence of blaKPC, blaNDM-1, blaIMP, and blaVIM. NA, PCR not performed.
A subset of 60 specimens from the point prevalence study were inoculated onto both chromID CARBA and HardyCHROM Carbapenemase to directly compare the performance of the two media using clinical specimens. After 24 h, 2 isolates, P. aeruginosa and Pseudomonas sp., were recovered on chromID CARBA (96.7% overall specificity; 100% specificity based upon direct interpretation of the chromogenic medium). From the same 60 specimens, 12 Enterococcus spp. were recovered on HardyCHROM carbapenemase (80% overall specificity; 100% specificity based upon direct interpretation of the chromogenic medium). After 48 h of incubation, 10 additional isolates were recovered on chromID CARBA, decreasing the overall specificity to 80%. An additional 18 isolates were recovered on HardyCHROM Carbapenemase after 48 h of incubation, resulting in an overall specificity of 50%.
Three of the 9 Enterobacteriaceae isolates recovered on chromID were susceptible to all carbapenems tested (Table 3). In addition, 40 of 50 P. aeruginosa isolates recovered were susceptible to imipenem and meropenem, while 10 isolates were nonsusceptible to both of these carbapenems.
TABLE 3.
Antimicrobial susceptibility profiles of Enterobacteriaceae recovered on chromID CARBA
| Isolate | Antimicrobial susceptibility toa: |
Carbapenemase PCR resultb | CTX-M PCR result | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| CAZ | CRO | FEP | TZP | IMI | MEM | ERT | DOR | ATM | |||
| Citrobacter freundii | S | S | S | S | S | S | S | S | S | Negative | Negative |
| Citrobacter sedlakii | R | R | R | R | S | S | S | S | R | Negative | Negative |
| Enterobacter cloacae complex | R | R | S | I | S | S | R | S | R | Negative | Negative |
| Enterobacter cloacae complex | R | R | S | R | S | I | R | I | R | Negative | Negative |
| Enterobacter sp. | R | R | S | S | S | S | I | S | R | Negative | Negative |
| Escherichia coli | R | R | R | R | I | S | S | S | R | Negative | Positive |
| Klebsiella oxytoca | R | R | R | R | S | S | S | S | R | Negative | Negative |
| Klebsiella oxytoca | S | R | S | I | S | S | S | S | S | Positive (KPC) | Positive |
| Serratia marcescens | R | R | R | R | I | R | R | S | R | Negative | Negative |
Antimicrobial susceptibility testing was performed using Kirby-Bauer disk diffusion. CAZ, ceftazidime; CRO, ceftriaxone; FEP, cefepime; TZP, piperacillin-tazobactam; IMI, imipenem; MEM, meropenem; ERT, ertapenem; DOR, doripenem; ATM, aztreonam; S, susceptible; R, resistant; I, intermediate.
Isolates were tested for the presence of blaKPC, blaNDM-1, blaIMP, and blaVIM.
Eight Enterobacteriaceae (n = 9), all P. putida (n = 3), and all carbapenem-nonsusceptible P. aeruginosa (n = 10) isolates were negative according to PCR for the blaKPC (13), blaNDM-1 (14), blaIMP, and blaVIM genes (15). One isolate of Klebsiella oxytoca was positive for the blaKPC gene. However, this isolate was susceptible to all four carbapenems as tested by disk diffusion. Additionally, the K. oxytoca isolate and the E. coli isolate were positive for the blaCTX-M gene (15).
Our findings contrast with those of Papadimitriou-Olivgeris et al. (16), who recently published a report evaluating chromID CARBA for detection of CPE from rectal swabs at a hospital in Greece. Their study evaluated fewer specimens (n = 177) and included only patients in intensive care, and the recovery of CPE was significantly higher (89 CPE from 86 specimens). A single isolate of Enterobacter cloacae was the only Enterobacteriaceae isolate recovered on chromID CARBA that did not produce a carbapenemase. The sensitivity of chromID CARBA at the patient level was 96.5%, while the specificity was 91.2% based upon the direct interpretation of the chromogenic medium and increased to 100% when Gram stain characteristics were included in the interpretation algorithm.
Day and colleagues (17, 18) recently published two articles investigating the prevalence of CPE in Pakistan using chromogenic media. Both studies had fewer subjects than our study (n = 175 and 152) and had higher rates of CPE (18.3% and 8.6%). The reported sensitivities of the chromogenic media were 100% and 92% in these studies, and both studies demonstrated recovery of non-CPE Enterobacteriaceae and significant numbers of Pseudomonas spp. While the analytical performance characteristics of chromID CARBA are favorable in areas of high prevalence (16–18), the positive predictive value of this medium for screening decreases in low-prevalence settings.
Both chromID CARBA and HardyCHROM carbapenemase are subject to breakthrough with non-CPE organisms. The colony morphology of non-Enterobacteriaceae isolates was distinct from that of Enterobacteriaceae, with the exception of the Aeromonas isolate, which was burgundy and resembled E. coli. In addition to the breakthrough of non-Enterobacteriaceae, chromID CARBA is also subject to growth of non-CPE Enterobacteriaceae. As a result of the proportion of non-CPE isolated on this medium, we recommend that all Enterobacteriaceae be confirmed as carbapenemase producing before reporting the presence of a CPE in the specimen. This can add to the turnaround time of the testing. Although Enterobacteriaceae isolates were morphologically distinct from other species recovered on the agar, there were no morphological differences between CPE and non-CPE isolates to guide this analysis prior to the subsequent characterization of the isolates.
Overall, chromID CARBA has analytical performance characteristics for CPE superior to those of HardyCHROM Carbapenemase agar. During initial testing, LODs for chromID CARBA were typically a log or more sensitive than those observed for the HardyCHROM Carbapenemase medium. While it is unclear if this difference is clinically significant, due to a paucity of data on the typical fecal burden of CPE, it is not uncommon for laboratories to select a screening approach with the best analytical sensitivity. Based on our low prevalence, it is unlikely that routine screening for CPE would be justified or cost-effective at our hospital. Health care institutions may want to conduct prevalence studies to gain a snapshot of local or institutional CPE carriage and inform the need for screening and preventive measures to control the spread of these important pathogens in their institution.
ACKNOWLEDGMENTS
We thank Meghan Butler for her assistance as the honest broker for clinical specimen deidentification.
The chromID CARBA agar for the point prevalence study was provided by bioMérieux.
M.A.P. is a recipient of the bioMérieux Scholarship in Clinical Microbiology. C.-A.D.B. has received research support from bioMérieux.
REFERENCES
- 1.Nordmann P, Naas T, Poirel L. 2011. Global spread of carbapenemase-producing Enterobacteriaceae. Emerg Infect Dis 17:1791–1798. doi: 10.3201/eid1710.110655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Walsh TR. 2010. Emerging carbapenemases: a global perspective. Int J Antimicrob Agents 36 (Suppl 3):S8–S14. [DOI] [PubMed] [Google Scholar]
- 3.Munoz-Price LS, Poirel L, Bonomo RA, Schwaber MJ, Daikos GL, Cormican M, Cornaglia G, Garau J, Gniadkowski M, Hayden MK, Kumarasamy K, Livermore DM, Maya JJ, Nordmann P, Patel JB, Paterson DL, Pitout J, Villegas MV, Wang H, Woodford N, Quinn JP. 2013. Clinical epidemiology of the global expansion of Klebsiella pneumoniae carbapenemases. Lancet Infect Dis 13:785–796. doi: 10.1016/S1473-3099(13)70190-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Dortet L, Poirel L, Nordmann P. 2014. Worldwide dissemination of the NDM-type carbapenemases in Gram-negative bacteria. Biomed Res Int 2014:249856. doi: 10.1155/2014/249856. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.CDC. 2012. Laboratory protocol for detection of carbapenem-resistant or carbapenemase-producing Klebsiella spp. and E. coli from rectal swabs. CDC, Atlanta, GA: http://www.cdc.gov/hai/pdfs/labsettings/klebsiella_or_ecoli.pdf. [Google Scholar]
- 6.Manji R, Bythrow M, Branda JA, Burnham CA, Ferraro MJ, Garner OB, Jennemann R, Lewinski MA, Mochon AB, Procop GW, Richter SS, Rychert JA, Sercia L, Westblade LF, Ginocchio CC. 2014. Multi-center evaluation of the Vitek(R) MS system for mass spectrometric identification of non-Enterobacteriaceae Gram-negative bacilli. Eur J Clin Microbiol Infect Dis 33:337–346. doi: 10.1007/s10096-013-1961-2. [DOI] [PubMed] [Google Scholar]
- 7.Richter SS, Sercia L, Branda JA, Burnham CA, Bythrow M, Ferraro MJ, Garner OB, Ginocchio CC, Jennemann R, Lewinski MA, Manji R, Mochon AB, Rychert JA, Westblade LF, Procop GW. 2013. Identification of Enterobacteriaceae by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry using the Vitek MS system. Eur J Clin Microbiol Infect Dis 32:1571–1578. doi: 10.1007/s10096-013-1912-y. [DOI] [PubMed] [Google Scholar]
- 8.Westblade LF, Jennemann R, Branda JA, Bythrow M, Ferraro MJ, Garner OB, Ginocchio CC, Lewinski MA, Manji R, Mochon AB, Procop GW, Richter SS, Rychert JA, Sercia L, Burnham CA. 2013. Multicenter study evaluating the Vitek MS system for identification of medically important yeasts. J Clin Microbiol 51:2267–2272. doi: 10.1128/JCM.00680-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Rychert J, Burnham CA, Bythrow M, Garner OB, Ginocchio CC, Jennemann R, Lewinski MA, Manji R, Mochon AB, Procop GW, Richter SS, Sercia L, Westblade LF, Ferraro MJ, Branda JA. 2013. Multicenter evaluation of the Vitek MS matrix-assisted laser desorption ionization-time of flight mass spectrometry system for identification of Gram-positive aerobic bacteria. J Clin Microbiol 51:2225–2231. doi: 10.1128/JCM.00682-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.McElvania Tekippe E, Shuey S, Winkler DW, Butler MA, Burnham CA. 2013. Optimizing identification of clinically relevant Gram-positive organisms by use of the Bruker Biotyper matrix-assisted laser desorption ionization-time of flight mass spectrometry system. J Clin Microbiol 51:1421–1427. doi: 10.1128/JCM.02680-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ford BA, Burnham CA. 2013. Optimization of routine identification of clinically relevant Gram-negative bacteria by use of matrix-assisted laser desorption ionization-time of flight mass spectrometry and the Bruker Biotyper. J Clin Microbiol 51:1412–1420. doi: 10.1128/JCM.01803-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.CLSI. 2013. Performance standards for antimicrobial susceptibility testing; 23rd informational supplement. CLSI document M100-S23. CLSI, Wayne, PA. [Google Scholar]
- 13.Doern CD, Dunne WM Jr, Burnham CA. 2011. Detection of Klebsiella pneumoniae carbapenemase (KPC) production in non-Klebsiella pneumoniae Enterobacteriaceae isolates by use of the Phoenix, Vitek 2, and disk diffusion methods. J Clin Microbiol 49:1143–1147. doi: 10.1128/JCM.02163-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.CDC. 2010. Multiplex real-time PCR detection of Klebsiella pneumoniae carbapenemase (KPC) and New Delhi metallo-β-lactamase (NDM-1). CDC, Atlanta, GA: http://www.cdc.gov/hai/pdfs/labsettings/kpc-ndm-protocol-2011.pdf. [Google Scholar]
- 15.Mangold KA, Voss BL, Singh K, Thomson RB Jr, Schora DM, Peterson LR, Kaul KL. 2013. Multiple broad-spectrum beta-lactamase targets for comprehensive surveillance. J Clin Microbiol 51:3423–3425. doi: 10.1128/JCM.01488-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Papadimitriou-Olivgeris M, Bartzavali C, Christofidou M, Bereksi N, Hey J, Zambardi G, Spiliopoulou I. 2014. Performance of chromID® CARBA medium for carbapenemases-producing Enterobacteriaceae detection during rectal screening. Eur J Clin Microbiol Infect Dis 33:35–40. doi: 10.1007/s10096-013-1925-6. [DOI] [PubMed] [Google Scholar]
- 17.Day KM, Ali S, Mirza IA, Sidjabat HE, Silvey A, Lanyon CV, Cummings SP, Abbasi SA, Raza MW, Paterson DL, Perry JD. 2013. Prevalence and molecular characterization of Enterobacteriaceae producing NDM-1 carbapenemase at a military hospital in Pakistan and evaluation of two chromogenic media. Diagn Microbiol Infect Dis 75:187–191. doi: 10.1016/j.diagmicrobio.2012.11.006. [DOI] [PubMed] [Google Scholar]
- 18.Day KM, Salman M, Kazi B, Sidjabat HE, Silvey A, Lanyon CV, Cummings SP, Ali MN, Raza MW, Paterson DL, Perry JD. 2013. Prevalence of NDM-1 carbapenemase in patients with diarrhoea in Pakistan and evaluation of two chromogenic culture media. J Appl Microbiol 114:1810–1816. doi: 10.1111/jam.12171. [DOI] [PubMed] [Google Scholar]